Patentable/Patents/US-10360688
US-10360688

Fast multi-spectral image registration by modeling platform motion

PublishedJuly 23, 2019
Assigneenot available in USPTO data we have
Inventorsnot available in USPTO data we have
Technical Abstract

A method of predicting warps for image registration includes receiving warp data for a first band pair in an N-band system. The method includes fitting a warp model to the warp data to produce an offset profile for the first band pair, predicting a respective offset profile for at least one other band pair in the N-band system, and using the predicted respective offset profile to generate a warp for the at least one other band pair. The method also includes registering images of the at least one other band pair using the warp for the at least one other band pair. In another aspect, a method of predicting warps for image registration includes receiving warp data for m band pairs in an N-band system, wherein m≤(N−1) and m>2, to register images.

Patent Claims
12 claims

Legal claims defining the scope of protection, as filed with the USPTO.

1

1. A method of predicting warps for image registration comprising: receiving warp data for a first band pair in an N-band system; fitting a warp model to the warp data to produce an offset profile for the first band pair; predicting a respective offset profile for at least one other band pair in the N-band system; using the predicted respective offset profile to generate a warp for the at least one other band pair; and registering images of the at least one other band pair using the warp for the at least one other band pair, wherein fitting a warp model to the warp data includes producing a slope profile for the first band pair, wherein the method includes predicting a respective slope profile for the at least one other band pair in the N-band system, and wherein using the predicted offset profile to generate a warp includes using both the predicted offset and slope profiles to generate the warp.

2

2. A method as recited in claim 1 , wherein predicting a respective offset profile includes shifting and scaling the offset profile for the first band pair.

3

3. A method as recited in claim 2 , wherein predicting a respective offset profile includes estimating stick spacing for the N-bands of a focal plane array (FPA).

4

4. A method as recited in claim 3 , wherein shifting and scaling values of the offset profile for the first band pair are derived from the stick spacing of the bands in the first band pair and the bands in the at least one other band pair.

5

5. A method as recited in claim 1 , wherein using the predicted respective offset profile to generate a warp for the at least one other band pair includes using a respective optical distortion profile for each respective band pair.

6

6. A method of predicting warps of image registration comprising: receiving warp data for a first band pair in an N-band system; fitting a warp model to the warp data to produce an offset profile for the first band pair; predicting a respective offset profile for at least one other band pair in the N-band system; using the predicted respective offset profile to generate a warp for the at least one other band pair; and registering images of at least one other band pair using the warp for at least one other band pair, wherein predicting a respective offset profile includes shifting and scaling the offset profile for the first band pair, wherein predicting a respective offset profile includes estimating stick spacing for the N-band of a focal plane array (FPA), wherein shifting and scaling values of the offset profile for the first band pair are derived from the stick spacing of the bands in the first band pair and the bands in the at least one other band pair, wherein the shifting and scaling values for predicting the at least one other band pair, d and e, from the first band pair, f and g, are given as Δ ⁢ ⁢ y de ⁡ ( fg ) = y d + y e - y f - y g 2 , ⁢ r de ⁡ ( fg ) = y e - y d y g - y f , respectively, where {y d ,y e ,y f ,y g } denote spatial locations of bands (d,e,f,g) respectively along a y direction in a focal plane array.

8

8. A method of predicting warps for image registration comprising: receiving warp data form band pairs in an N-band system, wherein m≤(N−1) and m>1; fitting respective warp models to the warp data for each respective one of the m band pairs to produce respective offset profiles for the m band pairs; predicting m respective offset profiles for at least one band pair in the N-band system; combining all the m predicted offset profiles into a single optimal offset profile for the at least one band pair of the N-band system; using the optimal offset profile to generate a warp for the at least one band pair; and registering images of the at least one band pair using the warp for the at least one band pair.

9

9. A method as recited in claim 8 , wherein using the predicted respective offset profile to generate a warp for the at least one other band pair includes using a respective optical distortion profile for the at least one other band pair.

10

10. A method as recited in claim 8 , wherein predicting a respective offset profile includes estimating stick spacing for the N-bands of a focal plane array (FPA).

11

11. A method as recited in claim 10 , wherein the m offset profiles predicted for the at least one band pair of the N-band system include an estimation of uncertainty of each prediction.

12

12. A method as recited in claim 11 , wherein combining the m predictions into an optimal offset profile takes into account the uncertainty of each prediction.

13

13. A method as recited in claim 12 , wherein the optimal offset profile predicted is obtained as a weighted sum of individual m predictions, wherein weights are inversely proportional to respective uncertainty and normalized to sum to one.

Classification Codes (CPC)

Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.

Patent Metadata

Filing Date

April 7, 2017

Publication Date

July 23, 2019

Want to explore more patents?

Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.

Citation & reuse

Analysis on this page is generated by Patentable — an AI-powered patent intelligence platform. AI-generated summaries, explanations, and analysis may be reused with attribution and a visible link back to the canonical URL below. Patent abstracts and claims are USPTO public domain.

Cite as: Patentable. “Fast multi-spectral image registration by modeling platform motion” (US-10360688). https://patentable.app/patents/US-10360688

© 2026 Patentable. All rights reserved.

Patentable is a research and drafting-assistant tool, not a law firm, and does not provide legal advice. Documents we generate are drafts for review by a licensed patent attorney.